How PDF Compression Works and Why Some Files Barely Shrink
PDF compression mostly means re-encoding the images inside a file, so the result depends heavily on what the file contains.
Lossless vs Lossy Compression
Lossless compression removes redundant data without discarding any information, useful for text and structure but limited in how much it can shrink a file. Lossy compression, used for photos and scans, discards some visual detail in exchange for a much smaller file. Most of the size reduction in a compressed PDF comes from lossy image re-encoding, not from the surrounding text or layout.
Image Compression and Quality
A quality setting from 1 to 100 controls how much detail a lossy image encoder keeps, with a typical default around 60. Lower quality means smaller files and more visible artifacts, especially around sharp edges and fine text inside an image. Higher quality preserves detail but shrinks the file less.
DPI and Downsampling
Downsampling reduces the pixel dimensions of an oversized embedded image before re-encoding it, so a photo saved at a print resolution far beyond what the page needs gets scaled down to something more reasonable for on-screen or standard-print viewing. This step alone can cut file size significantly, independent of the quality setting.
Scanned vs Vector vs Text PDFs
A scanned PDF is essentially one large image per page, so compression has a large effect on its size. A text-only PDF made from a word processor has little or no embedded image data, so re-encoding images barely changes its size, since there is little to re-encode. A PDF with vector graphics, such as charts or line art, behaves like a text PDF for compression purposes because vector shapes are not raster images.
Quality vs Size Trade-off
There is no setting that shrinks a file with zero visual cost when the file is image-heavy. Dropping quality from 90 to 60 typically gives a noticeable size reduction with modest visible loss; going below 40 starts to show artifacts on text-heavy scans. The right setting depends on whether the output needs to look good on screen, print cleanly, or just meet a file size limit for email or upload.
Typical Results by Document Type
- A scanned book chapter or contract: often shrinks substantially, since it is mostly large images.
- A slide deck with screenshots and photos: shrinks noticeably, more if screenshots are larger than needed.
- A word-processor report with only text and simple tables: shrinks very little, since there is almost nothing to re-encode.
- A design file with embedded high-resolution photos: shrinks well, with downsampling doing most of the work.
Troubleshooting
If a file did not shrink much, it is likely text-only or vector-only, and compression genuinely has little to work with. If the result looks soft, the quality setting was too aggressive for that content, raise it and re-run. If a file is still too big for an email attachment after compression, consider Split PDF to send fewer pages at once, or check whether Grayscale PDF is acceptable for the content, since dropping color data can help on color scans.
Related Tools
Start with Compress PDF and adjust the quality slider to taste. Resize PDF changes page dimensions rather than compressing content. Compress Image is useful when working with the source photos before they go into a PDF. PDF to JPG exports pages as images if you need a compressed picture rather than a compressed PDF. For how page dimensions and pixel counts relate to dpi, see the PDF page sizes guide.